Simpleshot: Revisiting Nearest-neighbor Classification For Few-shot Learning

Abstract

Few-shot learners aim to recognize new object classes based on a small number of labeled training examples. To prevent overfitting, state-of-the-art few-shot learners use meta-learning on convolutional-network features and perform classification using a nearest-neighbor classifier. This paper studies the accuracy of nearest-neighbor baselines without meta-learning. Surprisingly, we find simple feature transformations suffice to obtain competitive few-shot learning accuracies. For example, we find that a nearest-neighbor classifier used in combination with mean-subtraction and L2-normalization outperforms prior results in three out of five settings on the miniImageNet dataset.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keywang2019simpleshot

Related papers

• Object-level Representation Learning For Few-shot Image Classification (2018)0.00
• P3dc-shot: Prior-driven Discrete Data Calibration For Nearest-neighbor Few-shot Classification (2023)6.34
• Revisiting Metric Learning For Few-shot Image Classification (2019)14.90
• Finding Significant Features For Few-shot Learning Using Dimensionality Reduction (2021)2.26
• Bsnet: Bi-similarity Network For Few-shot Fine-grained Image Classification (2020)20.42
• Susana Distancia Is All You Need: Enforcing Class Separability In Metric Learning Via Two Novel Distance-based Loss Functions For Few-shot Image Classification (2023)0.00
• Learning From One And Only One Shot (2022)5.24
• Few-shot Learning Through An Information Retrieval Lens (2017)0.00